Interviews were carried out with all users of the system interface. All 10 waiters declared that it is difficult to use the application interface. They said that it is easy to get confused with the items and that it is often difficult to read their description. The bar is a pub-style craft beer bar. It has 30 options of craft beers on tap, more than 100 options in bottles and several food options.
This is an application used by waiters to take customer orders by selecting the products available on the menu.
Many orders are being sent wrong. As a consequence,
customers are dissatisfied.
The report was made by a craft beers restaurant / bar and the goal is find the problem and propose a solution.
Discovering interface problems
XS Rest
Why is it difficult to find items when searching?
The most common that was heard in the interviews was:
“You can't really read what is written and you can't always quickly identify if the beer is from a bottle or tap. For example, some bottles are very similar in size to medium-sized beer glasses. The bottle is 330ml and the glass is 350ml. Sometimes on the run we click on the wrong size. ”
​
Tables Arrangement
I analyzed the systems of the following companies to understand how they work. Understand the positive and negative points of each one. The truth is that they are very similar.
Main problems
After placing the order in the app, it is difficult to find the item among the various options that appear.
Problems reading the description of each item presented in the search.
Send wrong order to the kitchen or bar due to the difficulty of reading the descriptions.
Customer dissatisfaction when receiving what was not requested
Identify which tables are indoors and which are outdoors
Benchmark
I analyzed the systems of the following companies to understand how they work. Understand the positive and negative points of each one. The truth is that they are very similar.
Solution
based on the information collected in the interviews with the bar staff and with the analysis of similar systems from other companies, it was possible to understand the problem and propose a solution.
​
There is no need to redesign the software or propose to use a new one. It was understood that the app has layout settings and with the necessary adjustments it will be possible to adapt it for the bar / restaurant. The default configuration does not work well for most businesses.
Organizing
The user interface is very generic and does not have a suitability for each type of business. I mean, a craft beer bar has exactly the same interface as a pizzeria, when in fact some small customizations could be made in order to increase efficiency in the attendance.
The list of options would have 2 columns and not 3. with this new space the item information would have more space.
Dividing the beer categories by size, it is possible to quickly identify and place the order. The beers served in the glasses become blue in different shades according to the size. Bottle beers are in shades of gray.
beer style - Belgium Dubbel
Size
from tap
or bottle
Price
350ml
TAP
3,50
Mean sardine
Mamburana -
Beer with suggar mapple, amburana, melon and basil.
Beer Style
Brewery Name
Beer name
and description
TAP
500 ml
3,50
dark blue for 500ml
200 ml
TAP
3,50
light blue for 200ml
350 ml
TAP
3,50
blue for 350 ml
650 ml
Bottle
3,50
Gray for all kinds of bottle beers
The New Interface
It was realized that in the interviews with users and also analyzing the records of best-selling products, that in the period from 5pm to 7pm on weekdays (happy houor) some beer options and food with promotional prices are almost all orders.
a shortcut with these products will be proposed in order to expedite orders. The idea was approved by all users.
Shortcuts
During the research, it was understood that some of the items on the menu were ordered much more frequently during happy hours. Shortcuts for these items have been added to simplify the process.
interface of XS REST
bar / restaurant app
​
In the last month before the proposed changes were implemented, there was a record of 108 orders submitted that were wrong.
​
In the first month after the change the number of wrong orders dropped to 17
take a look at:
